Can people stop there personal attacks.?
STOP NOW
Dave ,group owner.?
On Thu, 22 Sep 2022 at 16:07, Mikek < amdx@...> wrote: On Thu, Aug 4, 2022 at 01:19 AM, Dr. David Kirkby, Kirkby Microwave Ltd wrote:
A step-down transformer seems the most obvious way, but that requires a turns ratio of sqrt(50000)=224. Even with a single turn on the secondary, there will be too many turns on the primary for this to work at 150 MHz.
Hi David,
Have you any more thoughts on the above conundrum? The only solution I see is to have two or three transformers to cover the frequency range.
????????????????????????????????????????????????????? Mikek
-- Dr. David Kirkby, Kirkby Microwave Ltd, drkirkby@...Telephone 01621-680100./ +44 1621 680100 Registered in England & Wales, company number 08914892. Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
|
On Sat, Sep 24, 2022 at 10:10 AM, Jim Strohm wrote:
Polystyrene or PVC?
They were labeled Styrene, I never found them listed as polystyrene. ?They are harder to find now than the were 5 years ago. ?Here's a 3 in styrene pipe coupler, This company has the 6", but no picture, just a drawing. ?I found them at a small hardware store and bought 5. ????????????? Mikek
|
Mikek,
That lathe is only 400 lbs, unless you max it out with accessories.
Two big amplifiers would weigh more than the lathe.
And it's not that expensive, either. If I needed a lathe, this one or
the next larger one would be a very good choice. But like you, I also
need a bigger shop. Next to the bigger shack I also need.
73
Jim N6OTQ
toggle quoted message
Show quoted text
On Sat, Sep 24, 2022 at 12:16 PM Mikek <amdx@...> wrote:
On Sat, Sep 24, 2022 at 07:09 AM, Tam Hanna wrote:
He just opened the door to a new world of pain. Been there, done that. Greater Machines are a hell of their own.
I hope it can only get better, we had a Horror Fright mini lathe where I worked and it was a pain to work with, to small and had a built in taper among other problems.
I'm looking at probably the PM1022V, and at least 4 additional accessories.
But there's always another accessory you need or wish you had.
I either need to build a new shop or put some support under the floor in the one I have.
Mikek
|
Hi Mikek,
Attached is representative Q-Curve for constant diameter, 6", and
inductance, 240uH, at varying tpi values.
As tpi decreases, turn count increases, wire resistance increases, Q
decreases.
Wire loss, Q=X/Rs, is a function of f/sqrt(f) Q increase as frequency
increases.
Core loss, parallel resistance, and possibly Q meter parallel
resistance, is Q=Rp/X. Q decreases as frequency increases. Assuming Rp
is a constant.
John KN5L
toggle quoted message
Show quoted text
On 9/24/22 12:04 PM, Mikek wrote: On Sat, Sep 24, 2022 at 04:54 AM, John KN5L wrote:
Attached is a representative Q Curve for a 240uH inductor.
Blue trace is XL/Rs series resistance wire loss Q
I don't get why Q is going up, losses generally increase faster that XL
does, and Q is always lower above about 1MHz than below, (contrawound
coils are slight exception)
Skin effect and core loss increase with frequency and maybe proximity
effect, I'm not sure about that.
Green Trace is Rp/XL parallel resistance core loss Q
Clueless
Red trace is wire and core combined Q
Ya, I don't get that, losses should increase with frequency, both wire
and core.
Black XL trace is added for XL math check.
Well, I get that one!
In above, core loss is a constant Rp. Based on your plot, core Rp
increases with frequency.
But I don't have a core, it's an air inductor, wound on a 6" polystyrene
pipe coupler.
Feedback?
????????????????????????????????? Mikek
|
On Sat, Sep 24, 2022 at 07:09 AM, Tam Hanna wrote:
He just opened the door to a new world of pain. Been there, done that. Greater Machines are a hell of their own.
?I hope it can only get better, we had a Horror Fright mini lathe where I worked and it was a pain to work with, to small and had a built in taper among other problems. I'm looking at probably the PM1022V, and at least 4 additional accessories. But there's always another accessory you need or wish you had. ?I either need to build a new shop or put some support under the floor in the one I have. ???????????????????????????????????????????????? Mikek
|
Polystyrene or PVC?
PVC is known to have issues as an inductor core.
REAL air-core inductors don't have coil forms, except maybe little
polystyrene stringers (as with Air-Dux) or thin paper, as with oatmeal
boxes for crystal radio sets.
73
Jim N6OTQ
toggle quoted message
Show quoted text
On Sat, Sep 24, 2022 at 12:04 PM Mikek <amdx@...> wrote:
On Sat, Sep 24, 2022 at 04:54 AM, John KN5L wrote:
Attached is a representative Q Curve for a 240uH inductor.
Blue trace is XL/Rs series resistance wire loss Q
I don't get why Q is going up, losses generally increase faster that XL does, and Q is always lower above about 1MHz than below, (contrawound coils are slight exception)
Skin effect and core loss increase with frequency and maybe proximity effect, I'm not sure about that.
Green Trace is Rp/XL parallel resistance core loss Q
Clueless
Red trace is wire and core combined Q
Ya, I don't get that, losses should increase with frequency, both wire and core.
Black XL trace is added for XL math check.
Well, I get that one!
In above, core loss is a constant Rp. Based on your plot, core Rp
increases with frequency.
But I don't have a core, it's an air inductor, wound on a 6" polystyrene pipe coupler.
Feedback?
Mikek
|
On Sat, Sep 24, 2022 at 04:54 AM, John KN5L wrote:
Attached is a representative Q Curve for a 240uH inductor.
Blue trace is XL/Rs series resistance wire loss Q
I don't get why Q is going up, losses generally increase faster that XL does, and Q is always lower above about 1MHz than below, (contrawound coils are slight exception) Skin effect and core loss increase with frequency and maybe proximity effect, I'm not sure about that.
Green Trace is Rp/XL parallel resistance core loss Q
Clueless
Red trace is wire and core combined Q
Ya, I don't get that, losses should increase with frequency, both wire and core.
Black XL trace is added for XL math check.
Well, I get that one!
In above, core loss is a constant Rp. Based on your plot, core Rp
increases with frequency.
But I don't have a core, it's an air inductor, wound on a 6" polystyrene pipe coupler. Feedback? ????????????????????????????????? Mikek
|
Hi,
>>>? Now just have to pull the trigger on that lathe
I've been looking at.
No need to be angry about him, Brian. He just opened the door to
a new world of pain. Been there, done that. Greater Machines are a
hell of their own.
With best regards
Tam HANNA
Enjoy electronics? Join 21k4 other followers by visiting the Crazy Electronics Lab at
On 2022. 09. 24. 8:53, Brian wrote:
toggle quoted message
Show quoted text
Hello
Mike,
I
suggest you exercise more care in posting personal insults.
In
most technical circles, it is considered standard practice
to label axes.
According
to the data sent with your first, unlabelled graph, where
you asserted Q of 1400 at 800 kHz, you have labelled this
one incorrectly.
Keep
this up, and you could be a worthy contributor to a wiki.
73
de Brian, VK2GCE
?
?
Your a pretty picky SOB aren't you Brian!
?I didn't like how that post was when I got done, it had a
drawing I
thought I'd deleted, and it split my text.
I don't like that you can't edit on this group. I have deleted
whole posts and
reposted before.
?I did the 5 coil experiment to see what the optimum turns
were for that
size coil.
When I get a lathe, I'll repurpose the 7 and 8 TPI to 11 and
13 TPI and rerun
the series.
It was a lot of work doing all the tests about 260
measurements in total, I ran
each frequency 4 times and
averaged them. Kind of curious if I'll get the same drop at
900kHz on 3 out of
the 5 coils.
?Now just have to pull the trigger on that lathe I've been
looking at.
????????????????????????????????????
Mikek
?
|
Hi Mikek,
Attached is a representative Q Curve for a 240uH inductor.
Blue trace is XL/Rs series resistance wire loss Q
Green Trace is Rp/XL parallel resistance core loss Q
Red trace is wire and core combined Q
Black XL trace is added for XL math check.
In above, core loss is a constant Rp. Based on your plot, core Rp
increases with frequency.
Check the equations
John KN5L
toggle quoted message
Show quoted text
On 9/24/22 5:10 AM, Mikek wrote: On Fri, Sep 23, 2022 at 11:53 PM, Brian wrote:
According to the data sent with your first, unlabelled graph, where
you asserted Q of 1400 at 800 kHz, you have labelled this one
incorrectly
? Yes, I got wrong, I even referenced the wrong post where the layout
form got messed up.
?????????????????????????????????????????????????????????? Mikek
|
On Fri, Sep 23, 2022 at 11:53 PM, Brian wrote:
According to the data sent with your first, unlabelled graph, where you asserted Q of 1400 at 800 kHz, you have labelled this one incorrectly
? Yes, I got wrong, I even referenced the wrong post where the layout form got messed up. ?????????????????????????????????????????????????????????? Mikek 
|
On Sat, 24 Sep 2022 at 04:20, Mikek < amdx@...> wrote: I don't like that you can't edit on this group. I have deleted whole posts and reposted before.
I have disabled editing of posts because I have found it more of a nuisance than benefit. Sometimes people edit posts to correct a typographical error. Then 100¡¯s or thousands of people get an email with the typo corrected.? It¡¯s even worse if their post bad multiple attachments.?
Dave (Group owner) -- Dr. David Kirkby, Kirkby Microwave Ltd, drkirkby@...Telephone 01621-680100./ +44 1621 680100 Registered in England & Wales, company number 08914892. Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
|
Hello Mike,
I suggest you exercise more care in posting personal insults. In
most technical circles, it is considered standard practice to label axes.
According to the data sent with your first, unlabelled graph, where
you asserted Q of 1400 at 800 kHz, you have labelled this one incorrectly.
Keep this up, and you could be a worthy contributor to a wiki.
73 de Brian, VK2GCE
?
?
Your a pretty picky SOB aren't you Brian!
?I didn't like how that post was when I got done, it had a drawing I
thought I'd deleted, and it split my text.
I don't like that you can't edit on this group. I have deleted whole posts and
reposted before.
?I did the 5 coil experiment to see what the optimum turns were for that
size coil.
When I get a lathe, I'll repurpose the 7 and 8 TPI to 11 and 13 TPI and rerun
the series.
It was a lot of work doing all the tests about 260 measurements in total, I ran
each frequency 4 times and
averaged them. Kind of curious if I'll get the same drop at 900kHz on 3 out of
the 5 coils.
?Now just have to pull the trigger on that lathe I've been looking at.
????????????????????????????????????
Mikek
?
|
That's the right idea. You might want to increase the capacitor by,
say, 10x or so to push the corner frequency down further than the
1.6kHz it is with 100pF.
I'm looking forward to seeing how it works out!
-- Cheers,
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 9/23/2022 20:28, Mikek wrote:
toggle quoted message
Show quoted text
On Fri, Sep 23, 2022 at 05:15 PM, Tom Lee wrote:
You should ac-couple the feedback instead of using
that 1k resistor, or else you run the risk of perturbing the
bias.
More like this?
????????? Mikek
_._,_._,_
|
On Fri, Sep 23, 2022 at 05:15 PM, Tom Lee wrote:
You should ac-couple the feedback instead of using that 1k resistor, or else you run the risk of perturbing the bias.
More like this? ????????? Mikek 
|
Your a pretty picky SOB aren't you Brian! ?I didn't like how that post was when I got done, it had a drawing I thought I'd deleted, and it split my text. I don't like that you can't edit on this group. I have deleted whole posts and reposted before. ?I did the 5 coil experiment to see what the optimum turns were for that size coil. When I get a lathe, I'll repurpose the 7 and 8 TPI to 11 and 13 TPI and rerun the series. It was a lot of work doing all the tests about 260 measurements in total, I ran each frequency 4 times and averaged them. Kind of curious if I'll get the same drop at 900kHz on 3 out of the 5 coils. ?Now just have to pull the trigger on that lathe I've been looking at. ???????????????????????????????????? Mikek ? 
|
Interesting graph, Mike,
?
But it really needs labels on the axes.
?
The ¡®observer effect¡¯ goes back to Lord Kelvin around 1860. Who,
as a scientist or engineer, relies on anybody¡¯s wiki? If you must quote, give
the proper reference. Text without context is mere pretext.
?
73 de Brian, VK2GCE
?
?
toggle quoted message
Show quoted text
On Fri, Sep 23, 2022 at 02:54 AM, John Kolb wrote:
I was thinking the capacitive divider was to reduce the C
loading of the voltmeter on the resonant circuit, but it's to reduce the input
resistance.
I remember a maxim, "when you measure something, you
change it" It is officially called the "Observer affect" from
the wiki, regarding electronics, (I think it is incomplete and needs more work)
"In ,
and
are usually wired in series or parallel to the circuit, and so by their very
presence affect the current or the voltage they are measuring by way of
presenting an additional real or complex
to the circuit, thus changing the and behavior of the circuit itself. Even a more passive device
such as a , which measures the wire current
without coming into physical contact with the wire, affects the current through
the circuit being measured because the ."
?My real world experience is with High Q coils for the AM band, about
240uh with Qs over 1400 at 800khz. This is an R of 1,600,000 at resonance.
putting a 10M¦¸ resistor in parallel lowers that to 1,379,310, this would cause
a 14% reduction in the Q measurement.
? The above measurements and coil are wound on 6" polystyrene pipe
couplers, TPI where cut on a thread cutting setting lathe.
<snip>
???????????????????????????????????
Thank, Mikek
|
You should ac-couple the feedback instead of using that 1k resistor,
or else you run the risk of perturbing the bias.
You reduce the effectiveness of bootstrapping by reducing its gain,
so don't employ that 1k/21k divider. Leave the 22k where it is. Add
a 1 megohm resistor in series with the 100M after lifting the
latter's lower leg from ground. Then connect a cap from the cathode
back to the junction of the 1M and 100M resistors. The value of the
cap will determine the low-frequency cutoff point, below which
bootstrapping becomes progressively less effective.
-- Cheers,
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 9/23/2022 17:06, Mikek wrote:
toggle quoted message
Show quoted text
On Fri, Sep 23, 2022 at 04:49 PM, Tom Lee wrote:
Bootstrapping would certainly boost the shunt
resistance at very low frequencies, but there's an inherent
grid-cathode conductance that grows quadratically
Here's my slightly educated Proposed Idea of how to do it in the
blue (on right side), but, I have other projects to finish before
I want to lift my 260A on the bench.
I'd be happy to see other ideas on component  values.
???????????????????????????????? Mikek
|
On Fri, Sep 23, 2022 at 04:49 PM, Tom Lee wrote:
Bootstrapping would certainly boost the shunt resistance at very low frequencies, but there's an inherent grid-cathode conductance that grows quadratically
Here's my slightly educated Proposed Idea of how to do it in the blue (on right side), but, I have other projects to finish before I want to lift my 260A on the bench. I'd be happy to see other ideas on component  values. ???????????????????????????????? Mikek
|
Bootstrapping would certainly boost the shunt resistance at very low frequencies, but there's an inherent grid-cathode conductance that grows quadratically with increasing frequency to spoil the fun fairly soon. At even low "rf" that conductance will greatly exceed 1/100megohms, so I am pessimistic that bootstrapping will buy much at frequencies of relevance for most uses. I'm used to seeing unbootstrapped numbers like 10s of kilohms at 10s of MHz for tubes that are probably not grossly different from the 535. So, I would guess that the law of diminishing returns kicks in pretty hard a bit above the audio band. A bit beyond that and bootstrapping is not going to help materially. And that assumes no additional conductance arising from ordinary leakage due to grunge, humid ambient, etc.
It's certainly worth a try, though, since the mod is quite easy (and easy to reverse).
Cheers
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
toggle quoted message
Show quoted text
On 9/23/2022 14:02, John Kolb wrote:
I wonder if the 260 voltmeter circuit could be bootstrapped by lifting the 100M resistor from ground, adding another resistor to ground, and adding a cap from the junction to the636A cathode?
John? KK6IL
On 9/23/2022 5:49 AM, Mikek wrote:
The Boonton 260A manual? has a section on 'Sources of Error' starting on Page 12.
<>
??I'm surprised that the the tube loading is 90M¦¸ at 50kHz but goes down from there until it is only 60k¦¸ at 50MHz.
?? I'm sure with modern electronic components and bootstraping, there is a much better circuit than the 535A tube voltmeter used in the 260A.
??Looking at the 4342A manual they say that the correction factor at 1MHz using the 513A Q standard is 1.04, that tells me that even their voltmeter input resistance is 9,300,000¦¸
??This is back to about the same 14% error measuring a 1400Q coil at 800kHz, i mentioned before.
?? Feel free to check my math and logic, my 513A Q standard is 253uh, indicated Q of 235 @1MHz. (effective Q of 251 with 7.9pf distributed capacitance)
??I'm here to learn, I've had a thing for Q and Q measuring for over 30 years, it fascinates me.
???????????????????????????????????? Thank, Mikek
|
On Fri, Sep 23, 2022 at 03:01 PM, Mikek wrote:
I wonder if the 260 voltmeter circuit could be bootstrapped by lifting the 100M resistor from ground, adding another resistor to ground, and adding a cap from the junction to the636A cathode?
? A little more reading and pondering and, I think bootstrapping only helps reduce the loading of the 100M¦¸ resistor and does nothing for the loading caused by the tube grid itself. ?On the other hand, I thought there are circuits that minimize gate capacitance of a fet... ?Discussion? ???????????????????????? Thanks Mikek
|
Dr. David Kirkby, Kirkby Microwave Ltd
Brian
